Key points are not available for this paper at this time.
The paper argues that it is a mistake to define habit as behaviour or action; as a regular conjunction of actions; as a stock; as a form of automaticity (although habit is acquired and activated automatically); as a tendency, propensity or disposition (even though habit acts tendentially); as a mechanism; and as a process (even though habit is acquired and activated via several processes). A taxonomic definition is provided wherein habit is a cognitive representation of a cue-action response.
Steve Fleetwood (Mon,) studied this question.